Female cynomolgus monkeys (Macaca fascicularis) were kept under 3 different housing conditions: individually in type A cages (45 X 45 X 60 cm), individually in type B cages (70 X 70 X 100 cm) and as couples in type B cages. Primigravida did not show early embryonic mortality, differing significantly from 11.5% early losses in multigravida. Early embryonic mortality was not affected by housing condition. Further reproductive failure rates did not differ significantly for primigravid (18.5%) and multigravid females (24.0%), though abortion tended to occur more frequently in primigravida. Perinatal mortality (16.1%) accounted for most of the losses under each housing condition. More successful pregnancies (90%) were recorded for females housed individually in type B cages than for females housed in type A cages (68%). About 50% of the couples originally established remained until weaning of their infants, yielding 77% viable offspring. For multigravid females statistical evaluation showed a significant effect of housing conditions on reproductive outcome (X2-test 0.01 less than P less than 0.05) that could be entirely attributed to low losses in females housed individually in type B cages. It is concluded that housing conditions can have a profound influence on reproductive success in cynomolgus monkeys.